Premium
SONHICA (Simple optimized non‐HIerarchical Cluster Analysis): A new tool for analysis of molecular conformations
Author(s) -
Bravi Gianpaolo,
Gancia Emanuela,
Zaliani Andrea,
Pegna Monica
Publication year - 1997
Publication title -
journal of computational chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.907
H-Index - 188
eISSN - 1096-987X
pISSN - 0192-8651
DOI - 10.1002/(sici)1096-987x(19970730)18:10<1295::aid-jcc4>3.0.co;2-i
Subject(s) - basis (linear algebra) , cluster analysis , hierarchical clustering , simple (philosophy) , similarity (geometry) , computer science , cluster (spacecraft) , algorithm , data mining , biological system , mathematics , artificial intelligence , geometry , image (mathematics) , biology , philosophy , epistemology , programming language
We describe a new clustering program, SONHICA (Simple OptimizedNon‐HIerarchical Cluster Analysis), developed to analyze large data sets ofmolecular conformations. Unlike traditional clustering methods, SONHICAdoes not make use of an overall index, like a distance, to evaluatesimilarity between objects. Each descriptor variable is comparedindividually on the basis of a preset threshold value. This assures highcontrol and sensitivity over the input variables. In addition, periodic andnonperiodic descriptors, such as dihedral angles and interatomic distances,can easily be used together. SONHICA generates clusters with the highestpossible density and all pairs of objects within a cluster are similar.These features make SONHICA particularly suitable for the analysis of datasets which tend to form globular clusters. This method was applied to theanalysis of a modified linear tetrapeptide, ITF1697, under investigationfor its anti‐ischemic properties, and a cyclic pentapeptide, BQ123, apotent antagonist of endothelin A. On the basis of the results presentedhere, SONHICA appears to be an interesting new tool in the field of theclustering methods applied to the analysis of molecular conformations.© 1997 John Wiley & Sons, Inc. J Comput Chem 18: 1295–1311, 1997